Hugging Face's logo

nbroad
/
longt5-base-global-mediasum

Summarization
Transformers
PyTorch
Safetensors
longt5
text2text-generation
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
5
longt5-base-global-mediasum / README.md
nbroad's picture
nbroad HF staff
add cnn results
c71ccbd
preview code
|
raw history blame
No virus
2.76 kB
metadata
license: cc-by-nc-sa-4.0
tags:
  - generated_from_trainer
  - longt5
  - summarization
model-index:
  - name: longt5-mediasum
    results:
      - task:
          type: summarization
          name: Summarization
        dataset:
          name: xsum
          type: xsum
          config: default
          split: test
        metrics:
          - name: ROUGE-1
            type: rouge
            value: 22.7044
            verified: true
          - name: ROUGE-2
            type: rouge
            value: 5.616
            verified: true
          - name: ROUGE-L
            type: rouge
            value: 18.0111
            verified: true
          - name: ROUGE-LSUM
            type: rouge
            value: 18.1554
            verified: true
          - name: loss
            type: loss
            value: 2.1656227111816406
            verified: true
          - name: gen_len
            type: gen_len
            value: 18.3527
            verified: true
      - task:
          type: summarization
          name: Summarization
        dataset:
          name: cnn_dailymail
          type: cnn_dailymail
          config: 3.0.0
          split: test
        metrics:
          - name: ROUGE-1
            type: rouge
            value: 21.1522
            verified: true
          - name: ROUGE-2
            type: rouge
            value: 8.1315
            verified: true
          - name: ROUGE-L
            type: rouge
            value: 16.6625
            verified: true
          - name: ROUGE-LSUM
            type: rouge
            value: 19.3603
            verified: true
          - name: loss
            type: loss
            value: 1.899269700050354
            verified: true
          - name: gen_len
            type: gen_len
            value: 17.853
            verified: true

longt5-mediasum

This model is a fine-tuned version of google/long-t5-tglobal-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.0129

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 12
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3.0

Training results

Training Loss Epoch Step Validation Loss
2.66 1.0 1667 2.0643
2.472 2.0 3334 2.0241
2.3574 3.0 5001 2.0129

Framework versions

  • Transformers 4.20.1
  • Pytorch 1.11.0a0+17540c5
  • Datasets 2.3.2
  • Tokenizers 0.12.1